Ten mother-infant dyads in which the mother abused drugs during pregnancy were compared to 10 matched drug-free dyads using a short form of the Parent-Child Early Relational Assessment to analyze videotaped mother-infant interactions. Interactions consisted of two five-minute segments: structured and unstructured play. Infants were 8-12 months of age. There was a consistent tendency for the drug abusing group mean scores to separate from the controls. Drug abusing dyads had significantly more (r = .71, Fisher's Exact Probability) ratings below 3.0 in the unstructured play situation for items that measured enthusiasm, responsivity to infant cues, and infant happiness. These categories provide preliminary evidence for those characteristics which may be most problematic in the relationship between drug abusing mothers and their infants, especially in situations in which mother is responsible for providing structure.